One of the largest sources of uncertainty in climate models comes from convective clouds
systems parametrizations, which are necessary at current horizontal resolutions, of about 25km.
In order to fully resolve the effects of such systems, horizontal resolutions need to
reach the 1-3km range, which entail a huge increase in computational costs.
In recent years, the performance of the world largest supercomputers has steadily
improved, to the point that we can now think of running a climate model
at cloud resolving resolution, with a Simulated Years Per Day (SYPD) throughput
that is suitable for decade or century long simulations.
However, the architectures of the newest supercomputers are becoming more and
more heterogeneous, which makes the task of keeping a performant code base more
and more challenging.
Here, we present our effort to upgrade the new non-hydrostatic atmosphere dycore of
E3SM to a version that is highly performant on a variety of architectures, including
GPUs, conventional CPUs, and many-core CPUs. When using GPUs, our implementation is
one of the fastest, achieving 0.97 SYPD on the NGPPS benchmark, when running on the full Summit supercomputer.